Pete Willemsen (pete-willemsen++at++uiowa.edu)
Sat, 15 Aug 1998 08:57:29 -0500
I looked into how I setup the pfSegSet and I'm telling it to cull
backfaces. In fact, here is how it is setup.
pfSegSet *tmp_seg_set;
tmp_seg_set = new pfSegSet();
tmp_seg_set->segs[0].pos.set(0.0f,0.0f,0.0f);
tmp_seg_set->segs[0].dir.set(0.0f,0.0f,-1.0f);
tmp_seg_set->segs[0].length = 10000.0f;
tmp_seg_set->activeMask = 1;
tmp_seg_set->mode = PFTRAV_IS_PRIM | PFTRAV_IS_NORM | PFTRAV_IS_CULL_BACK;
tmp_seg_set->isectMask = 0xffffffff;
tmp_seg_set->bound = NULL;
tmp_seg_set->userData = NULL;
tmp_seg_set->discFunc = NULL;
I have looked through when our database gets loaded and I didn't notice any
strange traversal masks being setup that might restrict intersection testing.
One intersesting experiment I tried was to copy the native version of
perfly from the Indigo2 Impact onto the Onyx2. When running on the Onyx2
using the database that's giving us trouble, the copied perfly's drive mode
places us at 0.0 always. However, when I run perfly native to the Onyx2,
drive mode works fine on this database. (To be complete: when running the
Indigo2 Impact's perfly on this database on the Indigo2, drive mode works
fine as well)
Any more ideas?
Very Confused,
Pete Willemsen
willemsn++at++cs.uiowa.edu
At 11:31 AM 8/14/98 -0700, Angus Dorbie wrote:
>Perhaps the problem is the face winding of the database polygons.
>Intersection tests will check face winding and eliminate back faces
>if told. This would be irrespective of the visible state information.
>
>It may be something more obvious like intersection traversal masks.
>
>Cheers,Angus.
>
>Pete Willemsen wrote:
>>
>> Hello Again.
>>
>> We have an application in which we use the pfSegSet to determine the
>> height and normal of the terrain. When we run this application on
>> either our Onyx iR (2RM) or our Indigo2 Impact the intersections work
>> fine for all of our databases.
>>
>> However, we recently had to run this application on an Onyx2 iR (2RM),
>> and so, to help the port, I statically linked our application to the
>> Performer libraries on our Indigo2 Impact.
>>
>> When we run one of our databases with this performer application on
>> the Onyx2, the pfHits and pfSegSet work fine (hits returning greater
>> than 0).
>>
>> But, when we run the same application with a different database, the
>> pfHit count returns 0.
>>
>> I'm confused by this behavior since the application and all databases
>> give the correct height and normal values when run on our Onyx and
>> Indigo2. Any ideas as to what might be causing this?
>>
>> (We do not have the option to recompile on the Onyx2 - also, just for
>> your knowledge, I did try our application with dynamic linking, but we
>> had the same results).
>>
>> Pete
>> ----------------------------------------------
>> Pete Willemsen Department of Comp. Sci.
>> willemsn++at++cs.uiowa.edu University of Iowa
>> ---
>> http://www.cs.uiowa.edu/~willemsn
>> ----------------------------------------------
>> =======================================================================
>> List Archives, FAQ, FTP: http://www.sgi.com/Technology/Performer/
>> Submissions: info-performer++at++sgi.com
>> Admin. requests: info-performer-request++at++sgi.com
>
>--
>"Only the mediocre are always at their best." -- Jean Giraudoux
>
>For advanced 3D graphics Performer + OpenGL based examples and tutors:
>http://www.dorbie.com/
>=======================================================================
>List Archives, FAQ, FTP: http://www.sgi.com/Technology/Performer/
> Submissions: info-performer++at++sgi.com
> Admin. requests: info-performer-request++at++sgi.com
>
>
Pete
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Pete Willemsen <pete-willemsen++at++uiowa.edu>
http://www.cs.uiowa.edu/~willemsn
Dept. of Computer Science, Univ. of Iowa
This archive was generated by hypermail 2.0b2 on Sat Aug 15 1998 - 06:58:15 PDT